Glycosylation mutants of animal cells have proved invaluable in delineating the pathways of cellular glycosylation mechanisms and in studying the biological consequences of specific changes in carbohydrate structure at the cell surface. We have previously isolated and partially characterized 11 different glycosylation mutants from Chinese hamster ovary (CHO) cells. These mutants were selected for resistance to the toxicity of plant lectins. They have been classified into eight recessive complementation groups and one dominant class. The major aim of this project is to use specific selection protocols designed to obtain the entire repertoire of lectin-resistant (LecR) glycosylation mutants of CHO cells. Sequential lectin selections which eliminate known LecR mutants and thereby select for new LecR phenotypes have been applied. Lectins other than those previously used have also been examined as selective agents. Lectins which interact with terminal sugar residues of carbohydrate chains are of particular interest since they might be expected to give rise to mutants expressing subtle structural alterations in surface carbohydrate. New isolates have been characterized genetically (by complementation analysis) and phenotypically for lectin-resistance properties. Seven new mutant types have been uncovered. Two belong to new recessive complementation groups, three behave dominantly and the remaining two represent either double mutants or intragenic complementation groups. All the CHO glycosylation mutants have been tested for tumorigenicity in nude mice. The majority form tumors similar to parental CHO. However, one mutant exhibits increased invasiveness and two mutants are essentially non-tumorigenic. Future experiments are aimed at determining whether surface carbohydrate plays a role in one or more of the many steps involved in tumor formation in this model system.